Parameter conversion method and system for multi-modal brain network atlas

Abstract

The invention provides a parameter conversion method and system for a multi-modal brain network atlas, and the method comprises the steps: obtaining and comparing a plurality of source brain regions and a plurality of target brain regions of a single subject, constructing an overlapping brain region set of each target brain region and the source brain region, counting the number of white matter fiber bundles and a brain function connection coefficient between the overlapped brain region sets, and respectively calculating remapping coefficients of a white matter fiber brain network and a brain function connection coefficient brain network on the basis of the number and the brain function connection coefficient; an overlapped brain region set of the target brain regions and the source brain regions of the multiple subjects and the corresponding white matter fiber bundle number and brain function connection coefficients are obtained and counted; and calculating the variance of the brain connection strength of the first experimental group and the first control group under the source map and the variance of the brain connection strength of the second experimental group and the second control group under the target map, and performing weighted summation on the source brain connection statistical magnitude between the overlapped brain region sets through the influence weight to obtain the target brain connection statistical magnitude between the target brain regions.

Description

Technical Field

[0001] The present invention relates to the technical field of brain network map conversion, and in particular to a parameter conversion method and system for multimodal brain network maps. Background Art

[0002] In neuroimaging research, brain white matter fiber networks and brain functional connectivity coefficient networks constructed based on magnetic resonance imaging (MRI) are widely used to analyze the connection relationships between different brain regions. Network maps can provide rich topological information and reveal differences in connection patterns between healthy individuals and disease groups. Current studies usually perform network analysis based on different brain maps. Each map has a unique partitioning scheme and applicable scenarios, allowing researchers to select the optimal map for analysis based on specific needs. However, due to the different regional division methods of different maps, it is difficult to directly compare the network characteristics between studies, affecting the feasibility of multicenter studies and meta-analysis.

[0003] In the prior art, different modal brain network maps are converted by matching regions of interest (ROI matching) or using interpolation methods; however, these methods lack standardized cross-atlas conversion methods, lack brain connectivity statistics conversion methods suitable for white matter fiber networks, and manual conversion errors are large. Network brain connectivity statistics under different atlases are difficult to directly convert, resulting in a lack of comparability between different research results. Failure to consider the degree of overlap between different atlas regions can lead to data loss or introduce additional errors. The lack of brain connectivity statistics conversion methods suitable for white matter fiber networks has, to a certain extent, limited the feasibility of multicenter studies and large-scale brain disease analysis based on magnetic resonance imaging data. Therefore, there is an urgent need for a standardized method that can accurately convert parameters between multimodal brain atlases. Summary of the Invention

[0004] In view of this, an embodiment of the present invention provides a parameter conversion method and system for multimodal brain network maps to eliminate or improve one or more defects existing in the prior art, and solves the problem in the prior art that parameters between various types of brain network maps of different modalities cannot be converted and directly compared and analyzed.

[0005] One aspect of the present invention provides a parameter conversion method for a multimodal brain network map, the method comprising the following steps:

[0006] Obtain multiple source brain regions in the source atlas and multiple target brain regions in the target atlas for multiple subjects, compare the source brain regions with the target brain regions, screen the source brain regions that overlap with each target brain region, and construct an overlapping brain region set for each target brain region;

[0007] When the source map and the target map are white matter fiber brain network maps, the process of individual map parameter conversion includes: counting the number of white matter fiber bundles in the source brain regions between each set of overlapping brain regions, and performing weighted summation of the number of white matter fiber bundles in the source brain regions between each set of overlapping brain regions based on a preset first remapping coefficient to obtain the number of white matter fiber bundles in the target brain regions between the target brain regions; the process of group map parameter conversion includes: determining the group to which the subject belongs according to the set factors, and obtaining a first experimental group and a first control group regarding the source map, and a second experimental group and a second control group regarding the target map; calculating the variance of the brain connection strength of the first experimental group, the first control group, the second experimental group, and the second control group respectively to measure the volatility difference and correcting each first remapping coefficient to obtain the influence weight on the target brain connection statistics; calculating the source brain connection statistics of each source brain region between the sets of overlapping brain regions, and performing weighted summation of the source brain connection statistics based on the influence weight to obtain the target brain connection statistics between the target brain regions;

[0008] When the source map and the target map are brain functional connectivity coefficient network maps, the process of individual map parameter conversion includes: counting the source brain region brain functional connectivity coefficients between each overlapping brain region set, and performing weighted summation of the source brain region brain functional connectivity coefficients between each overlapping brain region set based on the second remapping coefficient to obtain the target brain region brain functional connectivity coefficients between the target brain regions; the group map parameter conversion process includes: determining the group to which the subject belongs according to the set factors, and obtaining the first experimental group and the first control group regarding the source map, and the second experimental group and the second control group regarding the target map; calculating the variance of the brain connection strength of the first experimental group, the first control group, the second experimental group and the second control group respectively to measure the volatility difference and correcting each second remapping coefficient to obtain the influence weight on the target brain connection statistics; calculating the source brain connection statistics of each source brain region between the overlapping brain region sets, and performing weighted summation of the source brain connection statistics based on the influence weight to obtain the target brain connection statistics between the target brain regions;

[0009] Among them, the calculation process of the preset first remapping coefficient includes: obtaining multiple sample target brain areas of multiple sample subjects in the sample target map, and a set of sample overlapping brain areas corresponding to the sample target brain areas in the sample source map; counting the number of source brain area sample white matter fiber bundles between the set of sample overlapping brain areas corresponding to each sample target brain area, counting the number of target brain area sample white matter fiber bundles between each sample target brain area, weightedly summing the number of source brain area sample white matter fiber bundles and mapping them to the number of target brain area sample white matter fiber bundles, solving the weights to obtain independent mapping coefficients, and normalizing the independent mapping coefficients of multiple sample subjects to obtain the first remapping coefficient.

[0010] In some embodiments, the method includes solving the independent mapping coefficient using the least squares method based on the mapping relationship between the number of source brain region sample white matter fiber bundles between the sample overlapping brain region sets corresponding to each sample target brain region and the number of target brain region sample white matter fiber bundles between each sample target brain region, wherein the independent mapping coefficient k ij Satisfied mapping relationship expression:

[0011]

[0012] Among them, Y AB represents the number of target brain region white matter fiber bundles between target brain region A and target brain region B, X ij represents the number of white matter fiber bundles in the sample overlapping brain region set of the source brain region, p represents the number of source brain regions in the sample overlapping source brain region set of the sample target brain region A, and q represents the number of source brain regions in the sample overlapping source brain region set of the sample target brain region B;

[0013] Normalizing the independent mapping coefficients of multiple sample subjects to obtain the remapping coefficient, the remapping coefficient k ij * Satisfies the following expression:

[0014]

[0015] Among them, k m,ij Indicates the independent mapping coefficient when the sample subject number is m, X m,ij It represents the number of white matter fiber bundles in the source brain regions between the sets of overlapping brain regions when the sample subject number is m, and n represents the total number of samples.

[0016] In some embodiments, the number of white matter fiber bundles in the target brain region between the target brain regions is obtained by weighted summing the number of white matter fiber bundles in the source brain region based on the preset first remapping coefficient, and the number of white matter fiber bundles in the target brain region Y is obtained. EF Satisfies the following expression:

[0017]

[0018] Among them, Y EF k represents the number of white matter fiber bundles in the target brain area between the target brain area E and the target brain F, xy * represents the first remapping coefficient, X xy represents the number of white matter fiber bundles in the source brain regions between the overlapping brain region sets, s represents the number of source brain regions in the overlapping source brain region set of the target brain region E, and t represents the number of source brain regions in the overlapping source brain region set of the target brain region F.

[0019] In some embodiments, the target brain connectivity statistic T between target brain regions EF Satisfies the following expression:

[0020]

[0021] Where s represents the number of source brain regions in the set of overlapping source brain regions of the target brain region E, t represents the number of source brain regions in the set of overlapping source brain regions of the target brain region F, and L xy Indicates the influence weight, T xy represents the source brain connection statistics of each source brain region between the overlapping brain region sets, k xy * represents the remapping coefficient, represents the variance of brain connection strength of the first experimental group; represents the variance of brain connection strength in the first control group; represents the variance of brain connection strength in the second experimental group; represents the variance of brain connection strength in the second control group.

[0022] In some embodiments, the brain connection strength is represented by the number of white matter fiber bundles per unit area of ​​the brain; the more fiber bundles per unit area of ​​the brain, the greater the brain connection strength.

[0023] In some embodiments, based on the second remapping coefficient, a weighted sum of the source brain region brain functional connectivity coefficients between the overlapping brain region sets is performed to obtain the target brain region brain functional connectivity coefficients between the target brain regions, and the target brain region brain functional connectivity coefficients satisfy the following expression:

[0024]

[0025] k cd * =w c w d ;

[0026] Among them, R GH represents the target brain region brain functional connectivity coefficient between the target brain region G and the target brain H, represents the second remapping coefficient, r cd represents the functional connectivity coefficient of the source brain regions between the overlapping brain region sets, k represents the number of source brain regions in the overlapping source brain region set of the target brain region G, l represents the number of source brain regions in the overlapping source brain region set of the target brain region H, and w c represents the overlapping brain volume ratio coefficient in the set of overlapping brain regions of the target brain region G, w d Represents the overlapping brain volume ratio coefficient in the set of overlapping brain regions of the target brain region H.

[0027] In some embodiments, the method further comprises:

[0028] Call the convertStatistics function to input the source brain connection statistics, source atlas name, target atlas name, target atlas connection type, brain connection strength variance, file saving type of the target atlas after parameter conversion, and target atlas display type into the preset integrated development environment, and save and display the target atlas containing the number of white matter fiber bundles in the target brain area and the target brain connection statistics.

[0029] On the other hand, the present invention also provides a parameter conversion for a multimodal brain network map, including a processor, a memory, and a computer program / instruction stored in the memory, wherein the processor is used to execute the computer program / instruction, and when the computer program / instruction is executed, the system implements the steps of any of the above methods.

[0030] On the other hand, the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon, which implements the steps of any of the above methods when executed by a processor.

[0031] On the other hand, the present invention further provides a computer program product, comprising a computer program / instruction, which implements the steps of any of the above methods when executed by a processor.

[0032] The beneficial effects of the present invention are at least:

[0033] In the parameter conversion method and system for multimodal brain network maps described in the present invention, the source brain area and the target brain area are compared to obtain the degree of overlap between the source map and the target map and eliminate the brain area positioning error caused by modal differences between brain network maps of different modalities; the independent mapping coefficients calculated by multiple sample subjects are normalized to obtain remapping coefficients, and each remapping coefficient is corrected by the variance of the brain connection strength of the experimental group and the control group to obtain the influence weight on the target brain connection statistics, so that the parameter conversion process adapts to the characteristics of different subject groups and improves the accuracy and stability of parameter conversion; the number of white matter fiber bundles and brain connection statistics in the source map are converted to the number of white matter fiber bundles and brain connection statistics in the target map through the remapping coefficients and influence weights, thereby realizing parameter conversion between various brain network maps of different modalities and avoiding information loss during the parameter conversion process.

[0034] Additional advantages, objects, and features of the present invention will be set forth in part in the following description and will become apparent to those skilled in the art upon examination of the following or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained by the structures particularly pointed out in the description and drawings.

[0035] Those skilled in the art will understand that the purposes and advantages that can be achieved by the present invention are not limited to the above specific descriptions, and the above and other purposes that can be achieved by the present invention will be more clearly understood based on the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention. In the drawings:

[0037] Figure 1 Schematic diagram of the process of parameter conversion method for multimodal brain network map according to one embodiment of the present invention.

[0038] Figure 2 This is a structural diagram of a parameter conversion method for a multimodal brain network map according to an embodiment of the present invention.

[0039] Figure 3 Schematic diagram of the correlation between the true t-statistic and the target t-statistic before and after parameter conversion according to one embodiment of the present invention.

[0040] Figure 4 Schematic diagram of the correlation and retention ratio between the true Cohen's d and the target Cohen's d before and after parameter conversion according to an embodiment of the present invention.

[0041] Figure 5 Schematic diagram of the distribution of target Cohen's d after conversion to the DK-114 spectrum according to one embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0043] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, the accompanying drawings only show structures and / or processing steps closely related to the solutions according to the present invention, while other details that are not closely related to the present invention are omitted.

[0044] It should be emphasized that the term "include / comprises" when used herein refers to the existence of features, elements, steps or components, but does not exclude the existence or addition of one or more other features, elements, steps or components.

[0045] It should also be noted that, unless otherwise specified, the term "connection" herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0046] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0047] In the prior art, when converting white matter fiber brain networks across atlases, some studies convert them by matching regions of interest, or use interpolation methods for data mapping; however, these methods lack standardized cross-atlas conversion methods, lack parameter conversion methods, and manual conversion errors are large. Parameters under different atlases are difficult to convert directly, resulting in a lack of comparability between different research results. The lack of parameter conversion methods suitable for multimodal brain network atlases limits the feasibility of multicenter studies and large-scale brain disease analysis based on magnetic resonance imaging data; the present invention proposes a parameter conversion method and system for multimodal brain network atlases, which obtains and compares multiple source brain regions and multiple target brain regions of multiple subjects, and constructs overlapping brain region sets for each target brain region; when the source atlas and the target atlas are white matter fiber network atlases, individual atlas parameter conversion is performed, the number of source brain region white matter fiber bundles between each overlapping brain region set is counted, and the number of target brain region white matter fiber bundles between the target brain regions is obtained based on the first remapping coefficient; when the source atlas and the target atlas are brain functional connectivity coefficient network atlases, individual atlas parameter conversion is performed, the number of source brain region white matter fiber bundles between each overlapping brain region set is counted, and the number of target brain region white matter fiber bundles between the target brain regions is obtained based on the first remapping coefficient; The brain functional connectivity coefficient of the brain region is obtained based on the second remapping coefficient. When the brain functional connectivity coefficient network map and the brain white matter fiber network map are converted into group map parameters, the first experimental group and the first control group of the source map, and the second experimental group and the second control group of the target map to which the subject belongs are determined. The variance of the brain connection strength of each fiber bundle in each group is calculated and the influence weight on the target brain connection statistics is corrected for each remapping coefficient. The source brain connection statistics of each source brain region between the overlapping brain region sets are calculated. The target brain connection statistics between the target brain regions are obtained by weighted summation based on the influence weights; the calculation process of the first remapping coefficient includes: obtaining multiple sample target brain regions and corresponding sample overlapping brain region sets, counting the number of source brain region sample white matter fiber bundles between each sample overlapping brain region set and the number of target brain region sample white matter fiber bundles between each sample target brain region, solving multiple independent mapping coefficients that map the weighted sum of the number of source brain region sample white matter fiber bundles to the number of target brain region sample white matter fiber bundles, and normalizing them to obtain the first remapping coefficient.

[0048] Figure 1 Schematic diagram of a process for converting parameters of a multimodal brain network map according to an embodiment of the present invention. Specifically, the present application provides a method for converting parameters of a multimodal brain network map, which includes the following steps S101 to S103:

[0049] Step S101: Obtain multiple source brain regions in the source atlas and multiple target brain regions in the target atlas for multiple subjects, compare the source brain regions with the target brain regions, screen the source brain regions that overlap with each target brain region, and construct an overlapping brain region set for each target brain region.

[0050] Step S102: When the source map and the target map are white matter fiber brain network maps, the process of individual map parameter conversion includes: counting the number of white matter fiber bundles in the source brain regions between each overlapping brain region set, and performing weighted summation on the number of white matter fiber bundles in the source brain regions between each overlapping brain region set based on a preset first remapping coefficient to obtain the number of white matter fiber bundles in the target brain regions between the target brain regions; the group map parameter conversion process includes: determining the group to which the subject belongs according to the set factors, and obtaining a first experimental group and a first control group regarding the source map, and a second experimental group and a second control group regarding the target map; respectively calculating the variance of the brain connection strength of the first experimental group, the first control group, the second experimental group, and the second control group to measure the volatility difference and correcting each first remapping coefficient to obtain the influence weight on the target brain connection statistics; calculating the source brain connection statistics of each source brain region between the overlapping brain region sets, and performing weighted summation on the source brain connection statistics based on the influence weight to obtain the target brain connection statistics between the target brain regions.

[0051] Step S103: When the source map and the target map are brain functional connectivity coefficient brain network maps, the process of individual map parameter conversion includes: counting the source brain area brain functional connectivity coefficients between each overlapping brain area set, and performing weighted summation of the source brain area brain functional connectivity coefficients between each overlapping brain area set based on the second remapping coefficient to obtain the target brain area brain functional connectivity coefficients between the target brain areas; the group map parameter conversion process includes: determining the group to which the subject belongs according to the set factors, and obtaining the first experimental group and the first control group regarding the source map, and the second experimental group and the second control group regarding the target map; respectively calculating the variance of the brain connection strength of the first experimental group, the first control group, the second experimental group and the second control group to measure the volatility difference and correcting each second remapping coefficient to obtain the influence weight on the target brain connection statistics; calculating the source brain connection statistics of each source brain area between the overlapping brain area sets, and performing weighted summation of the source brain connection statistics based on the influence weight to obtain the target brain connection statistics between the target brain areas.

[0052] Among them, the calculation process of the preset first remapping coefficient includes: obtaining multiple sample target brain areas of multiple sample subjects in the sample target map, and a set of sample overlapping brain areas corresponding to the sample target brain areas in the sample source map; counting the number of source brain area sample white matter fiber bundles between the set of sample overlapping brain areas corresponding to each sample target brain area, counting the number of target brain area sample white matter fiber bundles between each sample target brain area, weightedly summing the number of source brain area sample white matter fiber bundles and mapping them to the number of target brain area sample white matter fiber bundles, solving the weights to obtain independent mapping coefficients, and normalizing the independent mapping coefficients of multiple sample subjects to obtain the remapping coefficients.

[0053] In step S101, the multimodal brain network map includes a brain white matter fiber brain network map and a brain functional connectivity coefficient network map. When the source map and the brain network map are converted into parameters, they belong to different types of maps of the same modality brain network map; the multimodal brain network map includes an automatic anatomical labeling map (AAL), a Desikan-Killian 114 map (DK114), a Schaefer 200 area functional map (Schaefer200), a human connectome project multimodal partitioning map (HCP_MMP), a Desikan-Killian map (DK), a Desikan-Killian 219 area map (DK219), a brain network group map (BN), an Aslan map (Arslan), Baldassano atlas, Brodmann atlas, Econommo atlas, independent component analysis atlas (Ica), neuroscience preprint network 500 atlas (Nspn500), Power functional network atlas (Power), Shen functional atlas (Shen), Schaefer 300 functional atlas (Schaefer300), Schaefer 400 functional atlas (Schaefer400); comparing the source atlas and target atlas of each subject, and obtaining the source brain regions contained in the corresponding positions of the source atlas at the positions of each target brain region and constructing the overlapping brain region sets of each target brain region. Further, the process of comparing the source brain regions and the target brain regions, screening the source brain regions overlapping with each target brain region, and constructing the overlapping brain region sets of each target brain region includes steps S1011 to S1013:

[0054] Step S1011: Check whether the shapes of the source map and the target map are consistent. If they are inconsistent, an abnormal prompt is issued and the parameter conversion process is stopped.

[0055] Step S1012: Read the numbers of multiple source brain regions in the source atlas and the numbers of multiple target brain regions in the target atlas using a preset atlas processing tool.

[0056] Step S1013: Compare the source brain region and the target brain region, screen the number of voxels in the source brain region that overlaps with each target brain region, and construct an overlapping brain region set for each target brain region using multiple source brain region numbers, multiple target brain region numbers, and the number of voxels in the overlapping source brain regions.

[0057] Specifically, checking whether the shapes of the source map and the target map are consistent is to ensure the consistency of the source brain area and the target brain area of ​​the subject when screening the source brain area overlapping with each target brain area; when the preset map processing tool reads the numbers of multiple source brain areas in the source map and the numbers of multiple target brain areas in the target map, illustratively, the preset map processing tool uses Python's nibabel and numpy tools; voxel is the smallest unit in the brain network map, which can be used to depict the complex morphology and internal structure of the brain. The higher the resolution of the brain network map, the higher the number of voxels; by counting the voxels in each area of ​​the brain network map, the brain's white matter fiber connections and functions can be quantified and used for research between individuals or groups.

[0058] In step S102, overlapping brain region sets of each target brain region in the white matter fiber brain network atlas are connected by fiber bundles, and the brain region neural information is transmitted and combined through the fiber bundles; the number of white matter fiber bundles reflects the strength of the connection between each brain region and affects the efficiency of brain region neural information transmission. According to the number of white matter fiber bundles, the topological structure between each brain region can be constructed to realize the research of brain networks and diagnosis of diseases of different subjects; the weighted summation of the number of fibers obtains the sum of the number of white matter fiber bundles between all overlapping brain region sets, and the first remapping coefficient maps the sum of the number of white matter fiber bundles between all overlapping brain region sets to the target atlas to obtain the number of white matter fiber bundles of the target brain regions between the target brain regions.

[0059] Furthermore, in the calculation process of the first remapping coefficient, the sample data set includes but is not limited to the HCP subject data set and the CHCP subject data set, the HCP subject data includes diffusion-weighted imaging (DWI) data of multiple sample subjects, and the diffusion-weighted imaging data is a type of image data obtained by magnetic resonance imaging (MRI) technology that reflects the diffusion motion characteristics of water molecules in the brain. The diffusion-weighted imaging data is presented in the form of a three-dimensional image, and each voxel contains the diffusion information of water molecules; the brain white matter fiber network atlas is based on the diffusion-weighted imaging data, reflecting the static physical connection of the brain anatomy; in some embodiments, the method includes using the least squares method to solve the independent mapping coefficient based on the mapping relationship between the number of source brain region sample white matter fiber bundles between the sample overlapping brain region sets corresponding to each sample target brain region and the number of target brain region sample white matter fiber bundles between each sample target brain region. The independent mapping coefficient k ij Satisfied mapping relationship expression:

[0060]

[0061] Among them, Y AB represents the number of target brain region white matter fiber bundles between target brain region A and target brain region B, X ij It represents the number of white matter fiber bundles of source brain regions between the sample overlapping brain region sets, p represents the number of source brain regions in the sample overlapping source brain region set of the sample target brain region A, and q represents the number of source brain regions in the sample overlapping source brain region set of the sample target brain region B.

[0062] Normalize the independent mapping coefficients of multiple sample subjects to obtain the first remapping coefficient, the first remapping coefficient k ij * Satisfies the following expression:

[0063]

[0064] Among them, k m,ij Indicates the independent mapping coefficient when the sample subject number is m, X m,ij It represents the number of white matter fiber bundles of source brain region samples between the sets of overlapping brain regions when the sample subject number is m, and n represents the total number of samples.

[0065] In some embodiments, the number of white matter fiber bundles in the target brain region between the target brain regions is obtained by weighted summing the number of white matter fiber bundles in the source brain region based on the preset first remapping coefficient, and the number of white matter fiber bundles in the target brain region Y is obtained. EF Satisfies the following expression:

[0066]

[0067] Among them, Y EF k represents the number of white matter fiber bundles in the target brain area between the target brain area E and the target brain F, xy * represents the first remapping coefficient, X xy represents the number of white matter fiber bundles in the source brain regions between the overlapping brain region sets, s represents the number of source brain regions in the overlapping source brain region set of the target brain region E, and t represents the number of source brain regions in the overlapping source brain region set of the target brain region F.

[0068] Furthermore, the white matter fiber brain network map and the brain functional connectivity coefficient network map use a consistent brain connection statistic conversion method when performing group map parameter conversion, and the setting factors include but are not limited to the health status, age, gender and occupation of the group; the first experimental group and the first control group respectively contain the source brain regions of the two groups, and the second experimental group and the second control group respectively contain the target brain regions of the two groups; the variance of the brain connection strength of the first experimental group and the first control group is the variance of the brain connection strength between the source brain regions in each overlapping brain region, and the variance of the brain connection strength of the second experimental group and the second control group is the variance of the brain connection strength between the target brain regions; in the white matter fiber brain network map, in some embodiments, the brain connection strength is represented by the number of white matter fiber bundles per unit area of ​​the brain region; the more fiber bundles per unit area of ​​the brain region, the greater the brain connection strength. In the brain functional connectivity coefficient brain network map, the brain connection strength is related to the ratio of overlapping brain volume, and the overlapping brain volume is the overlapping voxels of the overlapping brain region set and the target brain region. Exemplarily, the target brain connection statistic uses the t statistic to test whether the fluctuation difference of brain connection strength between two groups is significant, and the fluctuation difference of brain connection strength is obtained by calculating the variance; the source brain connection statistics of each source brain region between the overlapping brain region sets are converted into the target brain connection statistics between the target brain regions by influencing the weight; in the brain white matter fiber brain network atlas, in some embodiments, the target brain connection statistics T between the target brain regions EF Satisfies the following expression:

[0069]

[0070] Where s represents the number of source brain regions in the set of overlapping source brain regions of the target brain region E, t represents the number of source brain regions in the set of overlapping source brain regions of the target brain region F, and L xy Indicates the influence weight, T xy represents the source brain connection statistics of each source brain region between the overlapping brain region sets, k xy * represents the remapping coefficient, represents the variance of brain connection strength of the first experimental group; represents the variance of brain connection strength in the first control group; represents the variance of brain connection strength in the second experimental group; represents the variance of brain connection strength in the second control group.

[0071] In step S103, in some embodiments, a weighted sum of the source brain region functional connectivity coefficients between the overlapping brain region sets is performed based on the second remapping coefficient to obtain the target brain region functional connectivity coefficients between the target brain regions. The target brain region functional connectivity coefficients satisfy the following expression:

[0072]

[0073] k cd * =w c w d ;

[0074] Among them, R GH represents the target brain region brain functional connectivity coefficient between the target brain region G and the target brain H, represents the second remapping coefficient, r cd represents the functional connectivity coefficient of the source brain regions between the overlapping brain region sets, k represents the number of source brain regions in the overlapping source brain region set of the target brain region G, l represents the number of source brain regions in the overlapping source brain region set of the target brain region H, and w c represents the overlapping brain volume ratio coefficient in the set of overlapping brain regions of the target brain region G, w d Represents the overlapping brain volume ratio coefficient in the overlapping brain region set of the target brain region H. Specifically, the calculation process of the second remapping coefficient includes: obtaining multiple sample target brain regions of multiple sample subjects in the sample target atlas, and the sample overlapping brain region set corresponding to the sample target brain region in the sample source atlas; counting the source brain region sample brain volume between the sample overlapping brain region sets corresponding to each sample target brain region, counting the target brain region sample brain volume between each sample target brain region, obtaining the ratio coefficient of the overlapping voxels between each target brain region and the source brain region in the overlapping brain region set through the source brain region sample brain volume and the target brain region sample brain volume, and calculating the second remapping coefficient.

[0075] In some embodiments, the parameter conversion method for a multimodal brain network map further includes: calling the convertStatistics function to input the source brain connection statistics, source map name, target map name, target map connection type, brain connection strength variance, file save type of the target map after parameter conversion, and target map display type into a preset integrated development environment, saving and displaying the target map containing the number of white matter fiber bundles in the target brain region and the target brain connection statistics. Specifically, the method of the present invention runs in multiple programming language environments, including but not limited to Python or MATLAB, and the target map is saved as a CSV file and displayed as an SVG image.

[0076] On the other hand, the present invention also provides a parameter conversion for a multimodal brain network map, including a processor, a memory, and a computer program / instruction stored in the memory, the processor is used to execute the computer program / instruction, and when the computer program / instruction is executed, the system implements the steps of the above method.

[0077] On the other hand, the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon, which implements the steps of the above method when the program / instruction is executed by a processor.

[0078] On the other hand, the present invention also provides a computer program product, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0079] The present invention will be described below in conjunction with a specific embodiment:

[0080] Figure 2 This figure is a schematic diagram of a parameter conversion method for multimodal brain network maps according to one embodiment of the present invention. The present invention proposes a parameter conversion method for multimodal brain network maps (TACOS), which achieves parameter conversion between different modal brain network maps through a two-stage model calculation.

[0081] 1. The brain white matter fiber network map converts individual map parameters based on the connection of overlapping fibers.

[0082] (1) Calculate the regional overlap between the source map and the target map. Extract the spatial position of each brain region in the source map and the target map, and calculate the corresponding region set {a1, a2, ..., a p}, the number of brain regions corresponding to target map region A in the source map is p, and the corresponding region set {b1, b2, ..., b q}, the target atlas brain region B corresponds to q brain regions in the source atlas. The steps for calculating the corresponding region set for each target atlas brain region in the source atlas are as follows: use Python's nibabel and numpy tools to read NIFTI images; for each pair of target and source atlases, compare the overlapping regions of each brain region one by one, that is, the overlapping parts of voxels with the same label in the two atlases, and record the number of overlapping voxels; the source and target atlases are different forms of brain network atlases.

[0083] (2) Calculate the remapping coefficients of the fiber bundles at the individual level. On the HCP subject data, which contains the diffusion-weighted imaging data of multiple sample subjects, calculate the source atlas area a i and b j The number of white matter fiber bundles between ij , the number of white matter fiber bundles Y between target brain regions A and B AB , solve the independent mapping coefficient k ij , the expression is:

[0084]

[0085] Calculate the remapping coefficient at the group level The expression is:

[0086]

[0087] (3) Use the group-level remapping coefficient to perform remapping. The group-level remapping coefficient k is calculated by ij * Map the number of connections in the source graph to the target graph using the expression:

[0088]

[0089] 2. Brain functional connectivity coefficients The brain network map is converted into individual map parameters based on the proportion of overlapping brain volumes.

[0090] The time series of the target brain area A in the target atlas satisfies the following expression: Among them, V i Indicates a i Time series of brain regions. Region {a1,a2,...,a p} represents the source atlas region that spatially overlaps with the target brain region A, and the variable w i The target brain area A and the source brain area a i The scale coefficient of overlapping voxels between them is in the range of (0,1). The time series of the target brain region B in the target atlas satisfies the following expression: The correlation coefficient between target brain area A and target brain area B satisfies the following expression: Among them, R AB represents the correlation coefficient between brain area A and brain area B, Cov represents the covariance, and σ represents the standard deviation.

[0091] Due to spatial autocorrelation, the voxel time series in the same brain region show high correlation, so the time series U A and time series V i show a strong correlation, and their standard deviations are roughly equal. This similarity in standard deviations justifies the substitution in the equation, using V i Instead of U A , use V j Instead of U B , which yields the following results: (3-9) Among them, r ij Brain region a represents the spatial overlap between target brain region A and target brain region B i and b j The correlation coefficient of the specified connection between is the corresponding scaling factor for the number of overlapping voxels. In this method, the average ratio of the overlapping brain regions’ volumes is used instead of the voxel overlap ratio.

[0092] 3. The brain network map of white matter fibers and brain functional connectivity coefficients is converted into group map parameters based on the variance-weighted t-statistic.

[0093] (1) Calculate the connection variance map of the two groups of subjects on the source map. The two groups of subjects are the experimental group and the control group: calculate the variance of each connection in the experimental group (P) and the variance in the control group (C)

[0094] (2) Calculate the influence weight and convert the t statistic. The expression of the influence weight is:

[0095]

[0096] in, represents the variance of brain connection strength between regions A and B in the target atlas in the experimental group (P); represents the variance of brain connectivity strength between regions A and B in the target atlas in the control group (C); represents the variance of brain connection strength between regions A and B in the source map in the experimental group (P); represents the variance of brain connection strength between regions A and B in the source atlas in the control group (C); Represents the remapping coefficient at the group level.

[0097] Calculate the t statistic of the target map, the expression is:

[0098]

[0099] Among them, T AB represents the comprehensive t statistic between regions A and B in the target map; L ij represents the influence weight of the ijth pair of connections on the target area AB; T ij represents the t-statistic between region i and region j in the source map.

[0100] (3) After the above calculations, the t-statistic on the target map remains consistent with the source map and can be used for comparison and combined analysis of different studies.

[0101] 4. The parameter conversion for multimodal brain network maps in the present invention can be performed in brain network maps of different modalities; when the programming language is Python or MATLAB, by calling the convertStatistics function, the source brain connection statistics, source map name, target map name, target map connection type, brain connection strength variance, file saving type of the target map after parameter conversion, and target map display type are input into the preset integrated development environment to save and display the target map containing the number of white matter fiber bundles in the target brain area and the target brain connection statistics.

[0102] 5. Figure 3 Schematic diagram of the correlation between the true t-statistic and the target t-statistic before and after parameter conversion according to one embodiment of the present invention. The target t-statistic matrix shows the converted target t-statistics from the DK (n=68), DK-219 (n=219), HCP-MMP (n=360), BN (n=210), and Schaefer (n=200) maps to the DK-114 (n=114) map. The depth of the matrix points represents the magnitude of the target t-statistics. The correlation coefficient r represents the correlation between the target t-statistic and the true t-statistic, with the ordinate representing the target t-statistic and the abscissa representing the true t-statistic. The red dashed line in the figure represents the correlation between the target t-statistic after conversion using the present invention and the true t-statistic. The permutation test results show that the target t-statistic generated by the conversion according to the present invention is significantly higher than the null distribution of the correlation of the target t-statistics generated from other simulated conversions, with all p values ​​< 0.001. Figure 4 This is a schematic diagram of the correlation and retention ratio between the true Cohen's d and target Cohen's d before and after parameter conversion described in one embodiment of the present invention. The correlation is the Cohen's d correlation between the target Cohen's d mapping obtained by converting five different atlases and randomly mixing multiple atlases, and the true Cohen's d mapping in the DK-114 atlas. The Cohen's d mapping is an indicator used to measure the effect size, reflecting the difference in the mean values ​​and the degree of correlation between the two groups of parameters before and after the atlas conversion. The retention ratio diagram shows the proportion of connections that are still retained when the true effect size exceeds the set Cohen's d threshold; the ordinate represents the retention ratio, and the abscissa represents the Cohen's d threshold. The true effect size includes positive effect size and negative effect size. Figure 5 A schematic diagram of the target Cohen's d distribution after conversion to a DK-114 profile, as described in one embodiment of the present invention. This diagram illustrates the distribution of effect sizes after conversion corresponding to connections with a true effect size |Cohen's d| > 0.15. The ordinate represents the converted target Cohen's d, and the abscissa represents the profile type. Target effect sizes include both positive and negative effect sizes.

[0103] In summary, the present invention provides a parameter conversion method and system for multimodal brain network maps, which obtains and compares multiple source brain regions in a source map and multiple target brain regions in a target map for multiple subjects, screens the source brain regions overlapping with each target brain region and constructs a set of overlapping brain regions for each target brain region; when performing individual map parameter conversion of a brain white matter fiber network map, the number of white matter fiber bundles in the source brain regions between each set of overlapping brain regions is counted, and the number of white matter fiber bundles in the source brain regions is weightedly summed based on a preset remapping coefficient to obtain the number of white matter fiber bundles in the target brain regions between the target brain regions; the first remapping coefficient The calculation process includes: obtaining multiple sample target brain regions of multiple sample subjects in the sample target map, and a set of sample overlapping brain regions corresponding to the sample target brain regions in the sample source map; counting the number of source brain region sample white matter fiber bundles between the set of sample overlapping brain regions corresponding to each sample target brain region, counting the number of target brain region sample white matter fiber bundles between each sample target brain region, weightedly summing the number of source brain region sample white matter fiber bundles and mapping them to the number of target brain region sample white matter fiber bundles, solving the weights to obtain independent mapping coefficients, and normalizing the independent mapping coefficients of multiple sample subjects to obtain remapping coefficients. When converting individual map parameters for brain functional connectivity coefficient network maps, the source brain region functional connectivity coefficients between overlapping brain region sets are counted, and the target brain region functional connectivity coefficients between target brain regions are obtained by weighted summing the source brain region functional connectivity coefficients between overlapping brain region sets based on the second remapping coefficient. The group map parameter conversion process for two-modality brain network images includes: dividing the subjects into two groups according to the set factors, obtaining a first experimental group and a first control group for the source map, and a second experimental group and a second control group for the target map; calculating the variance of the brain connection strength of the first experimental group, the first control group, the second experimental group, and the second control group to measure the fluctuation difference, and modifying each remapping coefficient to obtain the influence weight on the target brain connection statistics. The source brain connection statistics of each source brain region between overlapping brain region sets are calculated, and the source brain connection statistics are weighted summed based on the influence weight to obtain the target brain connection statistics between target brain regions.

[0104] An embodiment of the present invention also provides a parameter conversion system for a multimodal brain network map, including a processor and a memory, wherein the processor and the memory can be connected via a bus or other means.

[0105] The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0106] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as the program instructions / modules corresponding to the method for shielding buttons on an in-vehicle display device in the embodiments of the present invention. The processor executes the non-transitory software programs, instructions, and modules stored in the memory to perform various processor functions and data processing.

[0107] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory may optionally include a memory remotely located relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0108] The one or more modules are stored in the memory, and when executed by the processor, perform the method described in this embodiment.

[0109] An embodiment of the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the art.

[0110] It should be understood by those skilled in the art that the various exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is specifically performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0111] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0112] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or replace features of other embodiments.

[0113] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations to the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A parameter conversion method for multimodal brain network maps, characterized in that: The method comprises the following steps: Obtain multiple source brain regions in the source atlas and multiple target brain regions in the target atlas for multiple subjects, compare the source brain regions with the target brain regions, screen the source brain regions that overlap with each target brain region, and construct an overlapping brain region set for each target brain region; When the source map and the target map are white matter fiber brain network maps, the process of individual map parameter conversion includes: counting the number of white matter fiber bundles in the source brain regions between each set of overlapping brain regions, and performing weighted summation of the number of white matter fiber bundles in the source brain regions between each set of overlapping brain regions based on a preset first remapping coefficient to obtain the number of white matter fiber bundles in the target brain regions between the target brain regions; the process of group map parameter conversion includes: determining the group to which the subject belongs according to the set factors, and obtaining a first experimental group and a first control group regarding the source map, and a second experimental group and a second control group regarding the target map; calculating the variance of the brain connection strength of the first experimental group, the first control group, the second experimental group, and the second control group respectively to measure the volatility difference and correcting each first remapping coefficient to obtain the influence weight on the target brain connection statistics; calculating the source brain connection statistics of each source brain region between the sets of overlapping brain regions, and performing weighted summation of the source brain connection statistics based on the influence weight to obtain the target brain connection statistics between the target brain regions; When the source map and the target map are brain functional connectivity coefficient network maps, the process of individual map parameter conversion includes: counting the source brain region brain functional connectivity coefficients between each overlapping brain region set, and performing weighted summation of the source brain region brain functional connectivity coefficients between each overlapping brain region set based on the second remapping coefficient to obtain the target brain region brain functional connectivity coefficients between the target brain regions; the group map parameter conversion process includes: determining the group to which the subject belongs according to the set factors, and obtaining the first experimental group and the first control group regarding the source map, and the second experimental group and the second control group regarding the target map; calculating the variance of the brain connection strength of the first experimental group, the first control group, the second experimental group and the second control group respectively to measure the volatility difference and correcting each second remapping coefficient to obtain the influence weight on the target brain connection statistics; calculating the source brain connection statistics of each source brain region between the overlapping brain region sets, and performing weighted summation of the source brain connection statistics based on the influence weight to obtain the target brain connection statistics between the target brain regions; Among them, the calculation process of the preset first remapping coefficient includes: obtaining multiple sample target brain areas of multiple sample subjects in the sample target map, and a set of sample overlapping brain areas corresponding to the sample target brain areas in the sample source map; counting the number of source brain area sample white matter fiber bundles between the set of sample overlapping brain areas corresponding to each sample target brain area, counting the number of target brain area sample white matter fiber bundles between each sample target brain area, weightedly summing the number of source brain area sample white matter fiber bundles and mapping them to the number of target brain area sample white matter fiber bundles, solving the weights to obtain independent mapping coefficients, and normalizing the independent mapping coefficients of multiple sample subjects to obtain the first remapping coefficient.

2. The parameter conversion method for multimodal brain network maps according to claim 1, characterized in that: The method includes solving the independent mapping coefficient using the least square method based on the mapping relationship between the number of source brain region sample white matter fiber bundles between the sample overlapping brain region sets corresponding to each sample target brain region and the number of target brain region sample white matter fiber bundles between each sample target brain region. The independent mapping coefficient k ij Satisfied mapping relationship expression: Among them, Y AB represents the number of target brain region white matter fiber bundles between target brain region A and target brain region B, X ij represents the number of white matter fiber bundles in the sample overlapping brain region set of the source brain region, p represents the number of source brain regions in the sample overlapping source brain region set of the sample target brain region A, and q represents the number of source brain regions in the sample overlapping source brain region set of the sample target brain region B; Normalizing the independent mapping coefficients of multiple sample subjects to obtain the remapping coefficient, the remapping coefficient k ij * Satisfies the following expression: Among them, k m,ij Indicates the independent mapping coefficient when the sample subject number is m, X m,ij It represents the number of white matter fiber bundles in the source brain regions between the sets of overlapping brain regions when the sample subject number is m, and n represents the total number of samples.

3. The parameter conversion method for multimodal brain network maps according to claim 2, characterized in that: The number of white matter fiber bundles in the target brain regions between the target brain regions is obtained by weighted summing the number of white matter fiber bundles in the source brain regions based on the preset first remapping coefficient. The number of white matter fiber bundles in the target brain regions Y EF Satisfies the following expression: Among them, Y EF k represents the number of white matter fiber bundles in the target brain area between the target brain area E and the target brain F, xy * represents the first remapping coefficient, X xy represents the number of white matter fiber bundles in the source brain regions between the overlapping brain region sets, s represents the number of source brain regions in the overlapping source brain region set of the target brain region E, and t represents the number of source brain regions in the overlapping source brain region set of the target brain region F.

4. The parameter conversion method for multimodal brain network maps according to claim 3 is characterized in that: Target brain connectivity statistics T between target brain regions EF Satisfies the following expression: Where s represents the number of source brain regions in the set of overlapping source brain regions of the target brain region E, t represents the number of source brain regions in the set of overlapping source brain regions of the target brain region F, and L xy Indicates the influence weight, T xy represents the source brain connection statistics of each source brain region between the overlapping brain region sets, k xy * represents the remapping coefficient, represents the variance of brain connection strength of the first experimental group; represents the variance of brain connection strength in the first control group; represents the variance of brain connection strength in the second experimental group; represents the variance of brain connection strength in the second control group.

5. The parameter conversion method for multimodal brain network maps according to claim 4 is characterized in that: The brain connection strength is represented by the number of white matter fiber bundles per unit area of ​​the brain; the more fiber bundles per unit area of ​​the brain, the greater the brain connection strength.

6. The parameter conversion method for multimodal brain network maps according to claim 1, characterized in that: Based on the second remapping coefficient, the brain functional connectivity coefficients of the source brain regions between the overlapping brain region sets are weighted summed to obtain the target brain region brain functional connectivity coefficients between the target brain regions. The target brain region brain functional connectivity coefficients satisfy the following expression: k cd * =w c w d ; Among them, R GH represents the target brain region brain functional connectivity coefficient between the target brain region G and the target brain H, represents the second remapping coefficient, r cd represents the functional connectivity coefficient of the source brain regions between the overlapping brain region sets, k represents the number of source brain regions in the overlapping source brain region set of the target brain region G, l represents the number of source brain regions in the overlapping source brain region set of the target brain region H, and w c represents the overlapping brain volume ratio coefficient in the set of overlapping brain regions of the target brain region G, w d Represents the overlapping brain volume ratio coefficient in the set of overlapping brain regions of the target brain region H.

7. The parameter conversion method for multimodal brain network maps according to claim 1, characterized in that: The method further comprises: Call the convert Statistics function to input the source brain connection statistics, source atlas name, target atlas name, target atlas connection type, brain connection strength variance, file saving type of the target atlas after parameter conversion, and target atlas display type into the preset integrated development environment, and save and display the target atlas containing the number of white matter fiber bundles in the target brain area and the target brain connection statistics.

8. A parameter conversion system for multimodal brain network mapping, comprising a processor, a memory, and a computer program / instruction stored in the memory, characterized in that: The processor is configured to execute the computer program / instructions. When the computer program / instructions are executed, the system implements the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images